Benchmark RadarAI BENCHMARK PROFILE
WaspMOT
A benchmark for long-term multi-object tracking of Trichogramma wasps in controlled ecological experiments, with 10 sequences of ~12,000 frames each and dense annotations. It evaluates identity preservation over extended durations.
- Released
- 2026-07-09
- Readiness
- Runnable
- Primary field
- General AI
Why it matters
Existing MOT benchmarks focus on short videos, which do not assess long-term identity preservation. This benchmark provides a controlled scenario with closed-set tracking, potentially revealing limitations in current methods that are not observable in conventional datasets.
Motivation
Multi-object tracking (MOT) has achieved strong performance on benchmarks dominated by short video sequences.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.